Malignant transformation is invariably accompanied by alterations in expression of integrin cell adhesion receptors. In this sense. integrins are tumor-associated antigens. However, it is not known by which mechanisms they may participate to tumor progression. A likely possibility is that they direct invasion and metastasis. We will investigate this issue in a breast cancer and in a keratinocyte transformation model, because of compelling evidence that deregulation of alpha6 integrins (alpha6Beta4 and alpha6Beta1) affects prognosis in breast cancer and keratinocyte tumors (i.e., skin and head-neck cancer). Normally, alpha6Beta4 integrins bind to a basement membrane laminin isoform, and promote formation of hemidesmosome anchoring sites. In contrast, alpha6Beta1 mediates transient adhesion to various laminin isoforms or endothelial cells. Our hypothesis is that a functional disequilibrium of integrins alpha6Beta4 and alpha6Beta1 critically contributes to the ability of carcinoma cells to cross basement membranes, migrate on laminin and attach to endothelial cells, setting the stage for invasion and metastasis. The breast cancer model comprises three human breast cancer cell lines which have distinct invasive and metastatic properties, and concomitant distinct patterns of alpha6 integrins expression. To establish causal connections, we will quantitate the ratio of alpha6Beta1 vs. alpha6Beta4 in these cells, and investigate its molecular basis by biochemical means, including cDNA cloning of alpha6 and Beta4 from each cell line to seek breast cancer- associated structural variants that may explain altered expression or functions (Aim 1). The significance to carcinogenesis of such variants or, in their absence, of the altered alpha6Beta1 vs. alpha6Beta4 expression levels will be evaluated by screening panels of breast cancer cells and tissues with variant- or heterodimer-specific probes (Aim 2). The alpha6 integrin profile of the three breast cancer cell lines will be modified by transfecting normal or variant alpha6 or Beta4, to determine whether overexpression, underexpression, gain or loss- of function mutations of the alpha6Beta1, the alpha6Beta4 heterodimer or both directly affect cell performance in in vitro and in vivo assays for adhesion, migration, invasion and metastasis (Aim 3). In the keratinocyte transformation model, we showed that overexpression of alpha6 causes papilloma-like lesions, and that inappropriate alpha6 isoforms occur in squamous carcinomas. We will alter systematically the alpha6 integrin profile of keratinocytes at distinct transformation stages by transfecting cDNAs of alpha6, Beta4 (or their breast cancer variants), and will test their invasive and metastatic properties in vitro with the assays of Aim 3, and in vivo by transplanting them in mice as skin grafts (Aim 4). By these combined approaches, we should be able to determine how central the functional disequilibrium of alpha6 integrins is to invasive or metastatic phenotype, and to find explanations for the alpha6 integrins deregulation observed in breast and keratinocyte cancers. These insights into the role of integrins in cancer may suggest new ways to approach epithelial cancers in humans.